JC virus (JCV) is a small DNA virus of the polyomavirus genus that is the agent of the subacute CNS demyelinating disease, progressive multifocal leukoencephalopathy (PML). PML is the cause of death in approximately 5% of AIDS patients in the USA. The closely related human polyomavirus, BK virus (BKV), infects primarily the kidney, and has been implicated in kidney transplant failure. JC virus is excreted in the urine of healthy individuals and infects a large percentage of the population worldwide. Much of our effort in this project have been directed at characterizing the natural variation in the DNA sequence that occurs in both the regulatory region and the coding region of the viral genome, and determining the significance of these rearrangements and mutations for both the biology of the virus, and the pathogenesis of CNS disease. JC virus has proved to be a useful marker of human population migrations because the virus probably co-evolved with the human species and different genotypes remain strongly associated with continental populations and their various ethnic groups. The regulatory region of virus excreted from the kidney (archetypal) is quite stable when circulating in the population, but rearranges extensively by unknown mechanisms when it attacks the brain of immunocompromised individuals. These brain-adapted variants are each unique in their regulatory regions, and fortunately, appear not to be regularly excreted in the urine. Analysis of JC virus strains in Papua New Guinea and the South Pacific, as well as among Native Americans in North America and Argentina, has helped to complete the picture of JC virus variation in human populations, and thus of the movements and interactions of various human populations over the past 100,000 years. Types 1 and 4 are European, Types 2 and 7 are Asian, and Types 3 and 6 are African. Type 5, previously thought to be a possible recombinant between Type 2B and Type 6, has now been shown in follow-up studies on additional urine samples from this same patient to be an ordinary Type 6 sequence. Thus, the supposed recombination event could not be confirmed, and this resistance to recombination remains one of the important advantages for study of JC virus as a viral marker in relation to population migration. Type 8 and Type 2E are newly characterized genotypes found in Papua New Guinea and Oceania. A genotype in Papua New Guinea (Type 8A) is confined to that island and may represent the earliest inhabitants of the region, whereas strains of more recent origin (Type 8B) are found throughout Melanesia. Throughout the Western Pacific there are strains designated Type 2E which, along with mainland strains known as Type 7A, may be associated with the Austronesian expansion beginning about 5,000 years ago. These Type 2E strains are most closely related to the Type 2A strains found in North Japan and Korea, and to the Native Americans of North and South America. The Native American strains are distinguished from those of Japan at only a few nucleotide positions, but do constitute a recognizable variant, Type 2A2. The relationship between Type 2A and Type 2E suggests a common ancestor in Northeast Asia for a South Pacific population, the Native Americans, and a portion of the Japanese and Korean populations. Further studies will help to elucidate Native American as well as Austronesian origins. Another focus of our research this year has been the regulatory region rearrangements occurring in disseminated BK virus infection during BKV-induced kidney allograft nephropathy (BKN). We have shown that the same BK virus rearrangement is found in the brain and CSF, and that this rearranged structure parallels exactly that of JC virus in the brain of PML patients. The common mechanism effecting these pathogenic rearrangements remains unknown. Elucidation of this mechanism may give important clues to both the fatal demyelinating infection by JC virus in the brain, and the rare dissemination of BK virus to other organs, including the brain. Monitoring of the strains of these viruses in circulation will be important to detect neurovirulent strains before they become a public health problem.